Profile-Guided Optimization (PGO) benchmark results

[zamazan4ik]

Hi!

Since the library cares a lot (according to the README) about performance, I decided to apply Profile-Guided Optimization (PGO) on it (as I already did for many other applications - check my GitHub repo). Hope the results will be interesting for someone.

Test environment

• Fedora 39
• Linux kernel 6.6.13
• AMD Ryzen 9 5900x
• 48 Gib RAM
• SSD Samsung 980 Pro 2 Tib
• Compiler - Rustc 1.75
• minitrace-rust version: the latest for now from the master branch on commit ee49263f56ebbe038ca70377166d319fdb170601
• Disabled Turbo boost (for more stable results across benchmark runs)

Benchmark

Built-in benchmarks are invoked with cargo bench --all-features --workspace. PGO instrumentation phase on benchmarks is done with cargo pgo bench -- --all-features --workspace. PGO optimization phase is done with cargo pgo optimize bench -- --all-features --workspace.

All PGO optimization steps are done with cargo-pgo tool.

Results

I got the following results:

• Release: https://gist.github.com/zamazan4ik/741d89a5e553433a50b7e74415b3748a
• PGO optimized compared to Release: https://gist.github.com/zamazan4ik/9a64b42e8618a3bc66c586ceae796d7e
• (just for reference) PGO instrumented compared to Release: https://gist.github.com/zamazan4ik/7ca47ae3aa8decf0a7cb1e1768996c7d

At least in the provided by the project benchmarks, I see measurable performance improvements in many cases.

Possible further steps

I can suggest the following things to consider:

• Perform more PGO benchmarks in other scenarios. If it shows improvements - add a note to the documentation about possible improvements in the tracing library performance with PGO (I guess somewhere in the README file will be enough).

I will be happy to answer all your questions about PGO.